Technology and engineering program

ICRAR is a science-driven body but a number of astronomy programs are linked closely to technology and engineering innovation.

This innovation will enable ASKAP and other Australian pathfinder science, and will contribute significantly to international SKA design efforts.

ICRAR has identified key engineering domain specialisations needed to deliver these outcomes; these skills are summarised in this section and form the basis of the Centre’s engineering recruitment program.

1. Sparse aperture arrays for the SKA
2. ASKAP science archive facility
3. High-performance computing for radio astronomy
4. Conceptual design for the SKA science computing system
5. High angular resolution radio astronomy
6. High-time-resolution radio astronomy
7. Murchison radio astronomy observatory (MRO) support

Back to top

Sparse aperture arrays for the SKA

The SKA Reference Design includes three sensor, or antenna, technologies: aperture phased array (AA), small dish plus phased array feed (SD+PAF), and small dish plus wideband single-pixel feed (SD+SPF).

Aperture arrays are all-electronic telescopes that confer advantages in terms of flexibility, scientific utility and, potentially, cost. Two variants can be distinguished: sparse AAs (inter-element spacing greater than a half-wavelength) and dense AAs (many elements per wavelength). Both have their merits, with sparse AAs offering greater effective collecting area for a given cost, but presenting the challenge of more complex calibration.

This ICRAR project, which is complementary to ASKAP, is aimed at undertaking research and development on next-generation sparse arrays largely, but not entirely, in the context of the European Aperture Array Verification Program (AAVP) - a new program designed to feed technical insight to the PrepSKA SKA Design Study.

As well as bringing core ICRAR engineering and science interests into an internationally coherent program, and building on local experience gained with the MWA, the project forges strategic and bench-level links between Europe and WA, increasing the exposure of the MRO to a critical audience and underlining the radio astronomy engineering collaborative capacity in the State.

Back to top

ICRAR science archive facility

ASKAP and MWA data from MRO will be turned into first order data products (images, cubes, tables) from the raw visibility data. These science data products will need to be captured, securely stored, classified and made available to science teams, PIs and the international astronomical community. ASKAP and MWA have the capable of producing 5 Petabytes of processed science data products per year making MRO one of the most data intensive observatories in astronomy and comparable to the LSST project in the optical.  ICRAR has an opportunity to design, build and eventually operate the ICRAR Science Archive Facility. The MRO SAF project would:

• Build on existing experience at ICRAR;
• Highlight and develop ICRAR’s credentials in data intensive science and technology
• Create a platform for collaboration with industry and other data intensive projects;
• Provide experience and prototypes necessary to design, build and operate an SKA-scale SAF;
• Provide ICRAR scientists with high quality access to ASKAP data, supported by ICARA high performance computing skills.

Back to top

High-performance computing for radio astronomy

The data flow and data processing needs of ASKAP and the SKA will require the development of specialised systems for rapid data capture, data transport and data manipulation. These high performance systems, will need to be scalable, cost-effective and power-efficient.

Recent developments in graphics processor technologies (GPUs) show many desirable attributes and opportunities for radio astronomy applications. ICRAR will develop a program in GPU and hybrid computing hardware and software to support ICRAR research in the transient uiverse, high spatial resolution astronomy and data-intensive science.

This work will also form part of the design studies for SKA science data system hardware, with particular emphasis on:

• power dissipation-to-performance ratios
• storage and computing architectures for radio astronomy specific algorithms and survey science needs
• new programming paradigms and hybrid computing
• high data rate applications.

Back to top

Conceptual design for the SKA science computing system

The SKA will be an iconic project for both radio astronomy and ICT.

SKA-class processing and storage facilities will probably be the largest systems in the world by 2025. Scaling from exiting approaches and systems design implies costs for power, hardware and software engineering that may not be affordable within the existing SKA cost of 1.5 billion euros.

New architectures, algorithms and programming models need to be developed to meet SKA needs, building on the needs of projects like ASKAP and LSST. ICRAR is ideally placed to make a major contribution to the international effort to produce a costed conceptual design for the SKA Science Data System as part of the PrepSKA program.

Back to top

High angular resolution radio astronomy

Hardware for the acquisition of VLBI data from ASKAP will need to be designed and produced.

To provide maximum utility, a mobile data acquisition system for VLBI will be produced; this could also be used at ASKP or at other VLBI telescopes in WA, potentially New Norcia, Yarragadee, or a refurbished Carnarvon OTC telescope.

The mobile system will consist of a rack of equipment, including a hardware interface to the ASKAP digital system, a digital backend system from the INAF VLBI group and a Mark5b+ data recorder.

Back to top

High-time-resolution radio astronomy

As with VLBI, high-time-resolution observations with ASKAP will require low-level access to the complex sample time series and/or time resolution total power measurements, with real-time data processing pipelines, buffering of data. This will require:

• significant digital hardware engineering to interface to ASKAP
• development of power-efficient high performance computing (HPC) solutions for deployment at ASKAP site
• development of algorithms for data processing analogous to pulsar data processing
• exploration of novel data processing algorithms.

Back to top

Murchison radio astronomy observatory (MRO) support

Several areas of engineering and technology specialisation are identified as required to ensure ICRAR makes an research and development contribution complementary to those of other major SKA institutes, including CSIRO.

These “domain specialisations” are applicable to a range of projects and can be summarised as:

• antenna engineering, with an emphasis on phased array techniques
• radio-frequency (RF) engineering, with a focus on sub-1GHz technologies, highly-integrated systems, and low-cost, short-haul data transport
• digital systems engineering, emphasizing flexible field-programmable gate array (FPGA) techniques
• computer systems engineering, including high performance computing (HPC) solutions
• software engineering, encompassing HPC applications and software instrument development.

Back to top